DEVELOPMENTAL DEFECTS AT THE FORAMEN OVALE * BRADLEY M. PATrEK, PH.D. (From the Department of Anatomy, University of Mickigan Medical School, Ann Arbor, Mich.) Many individual instances of patent foramen ovale are already on record in the literature. These cases, however, do not seem to have been studied as a group, either with a view to differentiating the types of malformation encountered in this location, or ascer- taining for the different kinds of defects the possible range of their variation in extent. A long-standing interest in the normal and defective development of the heart has, through the generous co- operation of colleagues, brought to me for study more material of this type than one person would ordinarily encounter. Recently a leave of absence afforded the further opportunity of studying the specimens accumulated in a group of pathological institutes with records covering a total of over 500,000 autopsies. Naturally not all the congenitally defective hearts from these autopsies had been preserved, but the extensiveness and variety of the material avail- able was exceptional. Using drawings made directly from my own or museum specimens as a basis, and supplementing this material from a study of the literature, I have attempted to assemble a brief, but freely illustrated, survey of the defects that may be en- countered at the foramen ovale. Being not a clinician but an em- bryologist, I have approached the subject from a morphological standpoint. It is hoped, however, that the material may prove a useful foundation for those interested in attacking the clinical problems associated with such defects. LITERATURE Publications dealing with failure of the foramen ovale to close have been appearing for more than three centuries. Many of the papers are so old that their viewpoint has become almost un- intelligible to us of today. Botalli in 1565, for example, seized on cases exhibiting an open foramen ovale as offering an improve- ment on Galen's idea that the blood entered the left side of the heart from the right by way of spaces between the .trabeculae of the interventricular septum (Dalton, I884, p. I37). The weight * Received for publication November 3, 1937. '35 I136 PATTEN of Botalli's name behind this erroneous conception delayed for many years the acceptance of Servetus' contention that the passage of blood from right to left "does not take place through the median wall of the heart as commonly believed; but, by a grand defice, the refined blood is driven from the right ventricle of the heart, in a long course through the lungs." The language in which Servetus elaborated his ideas well indicates the curious mixture of keen observation and dogma that pervaded the work of this period. "By the lungs it (the blood) is prepared, assum ng a bright color. It is mingled with the inspired air and purged of its fulginous mat- ter by exiration and so at length the left ventricle of the heart attracts by its diastole the whole mixture, a suitable . material that . may become vital spirit." (Translation from Dalton, I884, p. 115-) Unfortunately the old papers are by no means the only ones in the literature that throw little light on the subject. Many com- paratively recent articles are but superficial descriptions of isolated cases. An idea of the frequency with which papers based on i or 2 cases appear in the literature may be gathered from the fact that in 205 references cited by Poynter (I9I9) only 225 cases are in- volved. Many of these were merely clinical diagnoses of "open foramen ovale" with no confirmation by autopsy. Among the enormous number of papers on the subject disappointingly few contain both a good clinical history of the case and an adequate record of the autopsy findings. Viewing the literature as a whole there seem to have been three factors primarily responsible for the often contradictory and un- satisfactory information it contains. First is the deep rooted tradi- tion that the foramen ovale closes immediately following birth. Thus, in the absence of other findings accounting for death, an open foramen ovale in a young infant is frequently unjustly ac- cused. This has led to much misapprehension as to both the frequency of occurrence, and the functional significance, of an un- closed foramen ovale during the neonatal period. There has long been ample evidence that the foramen ovale is not closed immedi- ately after birth, but that its closure is a gradual process spreading over most of the first year (Aleksieyeff, I9OI; Alvarenga, I869; Elsisser, I852; Hinze, I893; Patten, 1930, 1931; Scammon and Norris, I9I8). Familiarty with this fact would have eliminated DEVELOPMENTAL DEFECTS AT FORAMEN OVALE 137 from the literature many papers describing as instances of "ab- normal patency of the foramen ovale" conditions perfectly normal for the age at which they were observed. For example, a paper published comparatively recently in a well known medical journal is based on the heart of an infant that lived but 6 hours after birth. Death was attributed to an open foramen ovale and an undosed ductus arteriosus! A second cause of confusion commonly encountered in the litera- ture is the failure to distinguish between conditions in which the foramen ovale is adequately covered by a valve which is not com- pletely adherent to the septum, and conditions in which a struc- tural defect of the valve or the septum makes it impossible for the foramen to be functionally closed. Incomplete adhesion of the valvula to the septum, with a resulting "probe-patency," is so common that it must be regarded as a variant of the normal rather than as an abnormality. The combined figures of ten different observers compiled from over 4000 autopsies in which this condi- tion was an object of special attention show that probe-patency exists in one out of every four or five adult hearts (see Table I). As long as the valvula foraminis ovalis adequately overlaps the limbus fossae ovalis, probe-patency appears to be no functional handicap to an otherwise normal individual. The inclusion in the literature of a large number of cases where the "defect at the foramen ovale" was mere probe-patency has led to the impression that functionally significant defects in this region are much more common than is actually the case. Still a third underlying difficulty in arriving at any clear inter- pretation of the significance of defects at the foramen ovale is one that seems inherent in the entire subject of congenital defects of the heart. There appears to have been a sort of collector's instinct obsessing contributors to this field. The more bizarre and compli- cated the case, the more interest it appears to arouse. From either the practical or the scientific standpoint this is unfortunate. The clinical picture especially is most confusing when several defects co-exist in the same heart. The only hope of arriving at any sound interpretation of such cases lies in better understanding of the developmental conditions responsible for, and the clinical mani- festations of, uncomplicated cases of specific defects in which the major characteristics of the condition stand out unequivocally. 138 PATTEN To attempt to give a systematic survey of all the articles in a field where such a large proportion of the material is either anti- quated or uncritical would not be profitable. In the course of preparing this paper about 3000 references on congenital defects of the heart were culled. Some 300 of these purported to deal with an open foramen ovale. Even this burdensome list undoubtedly fails to constitute a complete bibliography, for the literature is scattered among journals dealing with clinical medicine, pathology, physiology, anatomy, embryology, and even general biology. It has, therefore, seemed wiser to dismiss the literature as a whole with the foregoing general comments and deal only with a rela- tively few selected references in connection with matters on which they were found helpful. THE DEVELOPMENT OF THE INTERATRIAL SEPTAL SYSTEM The growth processes leading toward the establishment of con- ditions as they appear in the heart of a newborn infant and the changes in the heart following birth are fairly well covered in the embryological literature (Born, I889; Keibel and Mall, I9IO; Mall, I912; Odgers, 1935; Patten, Sommerfield and Paff, 1929; Tandler, I912 and 1913; Waterston, I9I8). Much of this in- formation, however, is so widely scattered and so uncorrelated that it is not readily utilizable by those working in other fields. For this reason, and also for the sake of emphasizing certain points especially pertinent to an understanding of the defective conditions under discussion, the following brief summary of the normal prenatal and postnatal development of the interatrial septa is given. In the separation of the primitive common atrium into right and left chambers two septa are directly involved. These, on the basis of their sequential appearance, are commonly called septum primum and septum secundum. The partitioning process starts in very young embryos, indications of the formation of septum primum being recognizable as early as the 5th week * of develop- ment. Starting as a crescentic ridge on the dorsocephalic part of the atrial well, septum primum grows toward the atrioventricular canal (Text-Figs. I, A and 4, A). * Ages as here given are approximate fertilization ages, for "menstrual age" add 14 days. DEVELOPMENTAL DEFECTS AT FORAMEN OVALE '39 At about the same time that septum primum is making its ap- pearance, the first indications of the impending division of the original common atrioventricular canal into a right and a left chan- nel become evident. Two local thickenings, one dorsally, the other ventrally located, appear in the walls of the canal.